Agent skill

Flux 3 Archival Formats

by black-forest-labs in black-forest-labs/skills

A skill your agent uses when a FLUX 3 video needs a period or archival look.

MITAuto-check passedAI & LLM Engineering

Install Flux 3 Archival Formats

skills CLI
$ npx skills add black-forest-labs/skills --skill flux-3-archival-formats -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install black-forest-labs/skills flux-3-archival-formats --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/black-forest-labs/skills.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/flux-3-archival-formats .claude/skills/flux-3-archival-formats && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
flux-3-archival-formats
GitHub stars
128
Token cost
~1.4k tokens
SKILL.md length
684 words
Files
2
Skills in repo
10
Repo updated
First seen
Licence
MIT

At a glance

A skill your agent uses when a FLUX 3 video needs a period or archival look.

  • Works in 3 steps: Reduced resolution: "soft, visibly… → Tape artifacts: scan lines, chroma… → Bright-source behavior: "soft hazy bloom…
  • A FLUX 3 video needs a period
  • SKILL.md covers Format artifacts, Name the degradation, not only…, Templates and Prove a style word works: run…, plus 2 more sections
  • Instructions only: no scripts, shell commands, URLs or credentials in SKILL.md

What it does

Flux 3 Archival Formats is an agent skill from black-forest-labs/skills. Use when a FLUX 3 video needs a period or archival look. Covers naming a recording format instead of a mood, per-format artifacts, and templates.

Its SKILL.md is about 1.4k tokens, which your agent loads only when the skill is triggered. The skill folder holds 1 other file (for example `metadata.json`).

It sits in AI & LLM Engineering. The repository describes itself as: Official agent skills from Black Forest Labs for FLUX image and video generation — prompting guides and API integration patterns for Claude Code, Codex, and any… The licence is MIT.

When your agent uses it

  • A FLUX 3 video needs a period

Example prompts

  • “/flux-3-archival-formats”

Workflow steps

3 steps, taken from the first numbered list in SKILL.md.

  1. Reduced resolution: "soft, visibly low-resolution tape image," not just "VHS."
  2. Tape artifacts: scan lines, chroma bleed, head-switching band, tracking flicker.
  3. Bright-source behavior: "soft hazy bloom around the sun." A sharp anamorphic lens

What it can do on your machine

Read from SKILL.md and the folder at commit 8907d51. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Links to these hosts (documentation or services it may open):

    • docs.bfl.ai

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Flux 3 Archival Formats loads about 1.4k tokens when it runs. Until then it costs about 42 tokens; SKILL.md has 684 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~42
When it runs · the whole SKILL.md, loaded when a task matches
~1.4k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from black-forest-labs/skills at commit 8907d51, republished under its MIT licence (© black-forest-labs). 684 words, ~1,439 tokens.

Download SKILL.mdSave it as .claude/skills/flux-3-archival-formats/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
flux-3-archival-formats
description
Use when a FLUX 3 video needs a period or archival look. Covers naming a recording format instead of a mood, per-format artifacts, and templates.
metadata.author
Black Forest Labs
metadata.version
1.0.0
metadata.tags
flux, flux-3, bfl, text-to-video, archival, format, style

FLUX 3 Archival Formats

A period look is requested by naming the machine that recorded it, not the feeling it should produce. "Vintage," "nostalgic," and "retro" are adjectives: nothing in the output can confirm or refute them. A recording format is hardware with named physical artifacts, so a reviewer can check the frame for each one and say whether the request landed.

Use this for a single standalone shot, or for a montage that must read as one era. Route supplied stills or clips to flux-3-keyframes-continuation, and shot craft that has nothing to do with period to flux-3-cinematic-inserts.

Format artifacts

Each row is a checklist. Name the format and the era together, then request the artifacts explicitly; a reviewer confirms the look by finding them in the frame.

FormatEraArtifacts to request and verify
Super 81970sRounded film-gate corners, heavy warm grain, blooming highlights, dust and hair in the gate, occasional splice jump, slight gate weave
VHS camcorder1980s4:3 pillarbox, chroma bleed, horizontal scan lines, magenta head-switching band along the bottom of frame, tracking-error flicker, washed-out low-contrast color
16mm documentary1960sMuted green-leaning stock, handheld weave, visible print wear and scratches, shallow depth from a fast prime
Hi8late 1990sVideo-sharp image, barrel distortion, mild color noise in shadows, timecode-era consumer framing

Hi8 separates least from a modern digital look, because its era is the least stylistically distinct. Expect a weaker result there and lean on the other three when the brief allows a choice.

Name the degradation, not only the format

A format label alone is not enough. A bare request for "1980s VHS" returns a sharp, modern-looking image wearing period colors: the model reads the era as a color grade. Low fidelity has to be requested as fidelity, in the prompt, alongside the format name.

Three phrases carry most of the weight:

  1. Reduced resolution: "soft, visibly low-resolution tape image," not just "VHS."
  2. Tape artifacts: scan lines, chroma bleed, head-switching band, tracking flicker.
  3. Bright-source behavior: "soft hazy bloom around the sun." A sharp anamorphic lens flare is one of the strongest signals of a modern synthetic image, and models reach for it by default.

Templates

Replace the bracketed fields; leave the format language intact.

text
Super 8 home-movie footage from the 1970s. [SUBJECT AND ACTION] in [SETTING], warm
afternoon light. Heavy warm-toned film grain, rounded film-gate corners, blooming
highlights, dust and hair visible in the gate, slight gate weave, occasional splice jump.
Handheld, softly focused, saturated faded film color.
text
1980s VHS camcorder footage, soft and visibly low-resolution, 4:3 pillarboxed.
[SUBJECT AND ACTION] in [SETTING]. Heavy chroma bleed, horizontal scan lines, a magenta
head-switching band along the bottom of frame, tracking-error flicker, washed-out
low-contrast color. Bright sources bloom as a soft haze rather than a sharp flare.
text
16mm documentary footage from the 1960s. [SUBJECT AND ACTION] in [SETTING]. Muted
green-leaning film stock, handheld weave, visible print wear and vertical scratches,
shallow depth of field from a fast prime, available light only.
text
Late-1990s Hi8 camcorder footage. [SUBJECT AND ACTION] in [SETTING]. Video-sharp but
noisy image, barrel distortion at the frame edges, color noise in the shadows, flat
consumer-camcorder exposure.
Show full SKILL.md (313 more words)Show less

Prove a style word works: run a control probe

Before committing a style vocabulary, test that it changes the output at all. Hold one subject constant, generate it once per candidate format, and generate one control with no format language whatsoever. If the formats do not separate from the control and from each other, the vocabulary is inert and the wording is doing nothing.

The result to look for is an artifact the prompt did not ask for. In one probe on a fixed subject, a VHS request produced an unprompted head-switching band: a physical tape artifact, not a color grade, which is evidence the model holds a real representation of the format. The clean control came back bright, sharp, and modern, resembling none of the four. Use this probe for any new style claim before it ships in a guide.

Montage

Hold one format constant across every shot. Mixing Super 8 grain into one clip and VHS scan lines into the next reads as an error rather than a style, and artifacts from two formats inside a single clip are physically contradictory. Vary subject, light, and time of day instead; the format is the thing tying the sequence together.

Review

A shot passes when the named artifacts for its format are present in the frame, no artifact from a second format appears, bright sources bloom rather than flare, and the image does not read as a sharp modern capture wearing period colors. Reject and reprompt rather than accepting a clip that only fails the fidelity test; that failure is the most common one and the easiest to talk yourself out of seeing.

Period home-movie subjects carry a content risk worth stating: childhood scenes such as bath time or a pool jump can render a child unclothed. Discard the render, exclude the file, and record why, rather than reviewing it further.

Reference: docs.bfl.ai

© black-forest-labs, MIT. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in skills/flux-3-archival-formats of black-forest-labs/skills.

  • SKILL.md
  • metadata.json

Open the folder on GitHubat commit 8907d51

Compare with similar skills

Flux 3 Archival Formats next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Flux 3 Archival Formats compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Flux 3 Archival Formats this skillblack-forest-labs/skills128—~1.4kAutomated safety check: PassMIT
Yichen Asrmcncarl/yichen-skills4.4k—~780Automated safety check: PassCustom licence
Youtube FetcherJimmySadek/youtube-fetcher-to-markdown485—~3.1kAutomated safety check: PassMIT
Comfyui AnimatoolShiroEirin/comfyui-good-anima481—~4.6kAutomated safety check: PassGPL-3.0
Comfyui Agent Skill MieMieMieeeee/comfyui-agent-skill116—~3.9kAutomated safety check: PassApache-2.0
End Of Speech Integrationrapidaai/voice-ai745—~877Automated safety check: PassCustom licence

Similar skills

  • Yichen Asr

    mcncarl/yichen-skills

    逸尘自用的统一音视频转写入口,在 StepFun Step ASR 与火山引擎豆包 ASR 之间按输出需求、安全边界和可用状态路由。用于本地音频或视频的纯文本转写、时间戳、SRT 字幕、口播粗剪,以及转写前体检;用户明确指定服务商时不得静默切换。Use when a local audio or video file needs transcription and the correct…

    4.4k GitHub stars~780 tokensUpdated 7 days ago
    AI & LLM EngineeringAuto-check passed
  • Youtube Fetcher

    JimmySadek/youtube-fetcher-to-markdown

    Retrieve transcripts from YouTube, Instagram, TikTok, X, Vimeo and other video sites, summarize or analyze what was said (and shown on screen), or save an Obsidian-ready Markdown knowledge-base note…

    485 GitHub stars~3.1k tokensUpdated 4 days ago
    AI & LLM EngineeringAuto-check passed
  • Comfyui Animatool

    ShiroEirin/comfyui-good-anima

    Route ALL Anima image generation: validate Danbooru hard anchors, form visual brief, assemble English prompts and args, then load comfyui-manager for workflow execution.

    481 GitHub stars~4.6k tokensUpdated 3 mo ago
    AI & LLM EngineeringAuto-check passed
  • Comfyui Agent Skill Mie

    MieMieeeee/comfyui-agent-skill

    Agent skill for running registered ComfyUI workflows through a stable CLI, and for importing a user's own ComfyUI workflow into their private registry after review.

    116 GitHub stars~3.9k tokensUpdated yesterday
    AI & LLM EngineeringAuto-check passed
  • End Of Speech Integration

    rapidaai/voice-ai

    Add or modify end-of-speech integrations in assistant-api with strict separation from VAD internals.

    745 GitHub stars~877 tokensUpdated 4 days ago
    AI & LLM EngineeringAuto-check passed
  • Flux Image

    danielmeppiel/agentic-sdlc-handbook

    Generate images with FLUX models (Black Forest Labs) via inference.sh CLI.

    158 GitHub starsUsed in 1 repo~778 tokens
    AI & LLM EngineeringAuto-check passed

More from black-forest-labs/skills

All 10 skills in this repo
  • Flux 3 Keyframes Continuation

    black-forest-labs/skills

    A skill your agent uses when a FLUX 3 video must be built from supplied images or video.

    128 GitHub stars~827 tokensUpdated 1 mo ago
    Auto-check passed
  • Bfl API

    black-forest-labs/skills

    BFL FLUX API integration guide covering endpoints, async polling patterns, rate limiting, error handling, webhooks, and regional endpoints with Python and TypeScript code examples.

    128 GitHub starsUsed in 1 repo~2.5k tokens
    Auto-check: notes
  • Flux 3 Product Ads

    black-forest-labs/skills

    A skill your agent uses when building a finished product ad from FLUX 3 - shot design, voiceover, action-to-word sync, evidence-gated copy, deterministic assembly, and QC gates that catch clipped…

    128 GitHub stars~9.8k tokensUpdated 1 mo ago
    Auto-check passed
  • Flux 3 Prompt Doctor

    black-forest-labs/skills

    A skill your agent uses when diagnosing a FLUX 3 brief before generation.

    128 GitHub stars~904 tokensUpdated 1 mo ago
    Auto-check passed
  • Flux 3 Generate

    black-forest-labs/skills

    A skill your agent uses when submitting or polling a FLUX 3 video API job.

    128 GitHub stars~1.1k tokensUpdated 1 mo ago
    Auto-check passed
  • Flux 3 Audio Dialogue

    black-forest-labs/skills

    A skill your agent uses when directing FLUX 3 audio, dialogue, or voiceover.

    128 GitHub stars~584 tokensUpdated 1 mo ago
    Auto-check passed

Questions about Flux 3 Archival Formats

What does Flux 3 Archival Formats do?

A skill your agent uses when a FLUX 3 video needs a period or archival look. Flux 3 Archival Formats is an agent skill from black-forest-labs/skills. Use when a FLUX 3 video needs a period or archival look.

When should I use Flux 3 Archival Formats?

Flux 3 Archival Formats fits situations like: A FLUX 3 video needs a period.

How do I install Flux 3 Archival Formats in Claude Code?

Run `npx skills add black-forest-labs/skills --skill flux-3-archival-formats -a claude-code`. Or copy the skill folder (skills/flux-3-archival-formats in black-forest-labs/skills) into .claude/skills/flux-3-archival-formats in your project. Claude Code loads it when a task matches its description.

How do I install Flux 3 Archival Formats in Codex?

Run `npx skills add black-forest-labs/skills --skill flux-3-archival-formats -a codex`. Or copy the skill folder (skills/flux-3-archival-formats in black-forest-labs/skills) into .agents/skills/flux-3-archival-formats in your project. Codex loads it when a task matches its description.

Can I use Flux 3 Archival Formats in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add black-forest-labs/skills --skill flux-3-archival-formats -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/flux-3-archival-formats, .gemini/skills/flux-3-archival-formats, .github/skills/flux-3-archival-formats and .opencode/skills/flux-3-archival-formats in your project.

What does Flux 3 Archival Formats need to run?

SKILL.md names no scripts, command-line tools or credentials: Flux 3 Archival Formats is instructions for the agent only.

Does Flux 3 Archival Formats access the network?

SKILL.md names 1 domain. As links in the text: docs.bfl.ai. This is read from the text; nothing was executed.

Is Flux 3 Archival Formats safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Flux 3 Archival Formats use?

Flux 3 Archival Formats is published under the MIT licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Flux 3 Archival Formats use?

About 1.4k tokens (SKILL.md is roughly 5.8k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Flux 3 Archival Formats?

Skills that share tags, products or a category with Flux 3 Archival Formats: Yichen Asr (mcncarl/yichen-skills, 4.4k stars), Youtube Fetcher (JimmySadek/youtube-fetcher-to-markdown, 485 stars), Comfyui Animatool (ShiroEirin/comfyui-good-anima, 481 stars) and Comfyui Agent Skill Mie (MieMieeeee/comfyui-agent-skill, 116 stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Flux 3 Archival Formats?

black-forest-labs (a GitHub organization) maintains it in black-forest-labs/skills, which has 128 GitHub stars. The repository holds 10 skills in this directory. The repository was last updated on September 9, 2026.

Source: black-forest-labs/skills on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.